What's cooking (2019/12 #02)
Junio C Hamano committed
Dec 5, 2019 at 14:33 UTC
30a9c26739772d5e224f635e7f705dc30ecbedb7
1 file changed
+411
-750
whats-cooking.txt
+411
-750
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Dec 2019, #01; Mon, 2)
4
-X-master-at: 228f53135a4a41a37b6be8e4d6e2b6153db4a8ed
5
-X-next-at: e41a19f04e19db375f8e58b74c6cb9d3a1495be5
3
+Subject: What's cooking in git.git (Dec 2019, #02; Thu, 5)
4
+X-master-at: 083378cc35c4dbcc607e4cdd24a5fca440163d17
5
+X-next-at: f85fac6f6eedeb72469441dac97a7b59b027bf1f
6
7
-What's cooking in git.git (Dec 2019, #01; Mon, 2)
7
+What's cooking in git.git (Dec 2019, #02; Thu, 5)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with
@@ -12,10 +12,7 @@ Here are the topics that have been cooking. Commits prefixed with
12
'+' are in 'next'. The ones marked with '.' do not appear in any of
13
the integration branches, but I am still holding onto them.
14
15
-The second batch of topics are now in 'master'; this was a rather
16
-large update, so there might be some fallout. Many topics are still
17
-in 'next' to be merged to 'master' soonish, and many others are
18
-waiting to be in 'next'.
15
+The third batch of topics are now in 'master'.
16
17
You can find the changes described here in the integration branches
18
of the repositories listed at
@@ -25,417 +22,426 @@ of the repositories listed at
22
--------------------------------------------------
23
[Graduated to "master"]
24
28
-* ar/install-doc-update-cmds-needing-the-shell (2019-11-15) 1 commit
29
- (merged to 'next' on 2019-11-19 at 216468eb98)
30
- + INSTALL: use existing shell scripts as example
25
+* dd/rebase-merge-reserves-onto-label (2019-11-20) 1 commit
26
+ (merged to 'next' on 2019-11-21 at a6e4e5af0a)
27
+ + sequencer: handle rebase-merges for "onto" message
28
+ (this branch uses js/rebase-r-safer-label.)
29
32
- Doc update.
30
+ The logic to avoid duplicate label names generated by "git rebase
31
+ --rebase-merges" forgot that the machinery itself uses "onto" as a
32
+ label name, which must be avoided by auto-generated labels, which
33
+ has been corrected.
34
35
35
-* cb/curl-use-xmalloc (2019-11-07) 1 commit
36
- (merged to 'next' on 2019-11-19 at 1c587d0775)
37
- + remote-curl: unbreak http.extraHeader with custom allocators
36
+* dl/range-diff-with-notes (2019-11-21) 10 commits
37
+ (merged to 'next' on 2019-11-25 at dd6dad97c1)
38
+ + format-patch: pass notes configuration to range-diff
39
+ + range-diff: pass through --notes to `git log`
40
+ + range-diff: output `## Notes ##` header
41
+ + t3206: range-diff compares logs with commit notes
42
+ + t3206: s/expected/expect/
43
+ + t3206: disable parameter substitution in heredoc
44
+ + t3206: remove spaces after redirect operators
45
+ + pretty-options.txt: --notes accepts a ref instead of treeish
46
+ + rev-list-options.txt: remove reference to --show-notes
47
+ + argv-array: add space after `while`
48
39
- HTTP transport had possible allocator/deallocator mismatch, which
40
- has been corrected.
49
+ "git range-diff" learned to take the "--notes=<ref>" and the
50
+ "--no-notes" options to control the commit notes included in the
51
+ log message that gets compared.
52
53
43
-* dd/sequencer-utf8 (2019-11-11) 8 commits
44
- (merged to 'next' on 2019-11-19 at 9432ddc32c)
45
- + sequencer: reencode commit message for am/rebase --show-current-patch
46
- + sequencer: reencode old merge-commit message
47
- + sequencer: reencode squashing commit's message
48
- + sequencer: reencode revert/cherry-pick's todo list
49
- + sequencer: reencode to utf-8 before arrange rebase's todo list
50
- + t3900: demonstrate git-rebase problem with multi encoding
51
- + configure.ac: define ICONV_OMITS_BOM if necessary
52
- + t0028: eliminate non-standard usage of printf
54
+* dl/t5520-cleanup (2019-11-21) 15 commits
55
+ (merged to 'next' on 2019-11-27 at 523d920acf)
56
+ + t5520: replace `! git` with `test_must_fail git`
57
+ + t5520: remove redundant lines in test cases
58
+ + t5520: replace $(cat ...) comparison with test_cmp
59
+ + t5520: don't put git in upstream of pipe
60
+ + t5520: test single-line files by git with test_cmp
61
+ + t5520: use test_cmp_rev where possible
62
+ + t5520: replace test -{n,z} with test-lib functions
63
+ + t5520: use test_line_count where possible
64
+ + t5520: remove spaces after redirect operator
65
+ + t5520: replace test -f with test-lib functions
66
+ + t5520: let sed open its own input
67
+ + t5520: use sq for test case names
68
+ + t5520: improve test style
69
+ + t: teach test_cmp_rev to accept ! for not-equals
70
+ + t0000: test multiple local assignment
71
54
- Handling of commit objects that use non UTF-8 encoding during
55
- "rebase -i" has been improved.
72
+ Test cleanup.
73
74
58
-* dj/typofix-merge-strat (2019-11-11) 1 commit
59
- (merged to 'next' on 2019-11-19 at 5b6c0cf125)
60
- + merge-strategies: fix typo "reflected to" to "reflected in"
75
+* ec/fetch-mark-common-refs-trace2 (2019-11-20) 1 commit
76
+ (merged to 'next' on 2019-11-25 at c2b8e1546c)
77
+ + fetch: add trace2 instrumentation
78
62
- Typofix.
79
+ Trace2 annotation.
80
81
65
-* dl/complete-rebase-onto (2019-11-12) 1 commit
66
- (merged to 'next' on 2019-11-19 at ba53946a3b)
67
- + completion: learn to complete `git rebase --onto=`
82
+* ep/guard-kset-tar-headers (2019-11-07) 1 commit
83
+ (merged to 'next' on 2019-11-21 at 6c9574b01d)
84
+ + kset.h, tar.h: add missing header guard to prevent multiple inclusion
85
69
- The completion script (in contrib/) learned that the "--onto"
70
- option of "git rebase" can take its argument as the value of the
71
- option.
86
+ Code clean-up.
87
88
74
-* dl/doc-diff-no-index-implies-exit-code (2019-11-02) 1 commit
75
- (merged to 'next' on 2019-11-19 at e2b2181e28)
76
- + git-diff.txt: document return code of `--no-index`
89
+* hi/gpg-optional-pkfp-fix (2019-11-23) 2 commits
90
+ (merged to 'next' on 2019-11-25 at b0d668176f)
91
+ + gpg-interface: limit search for primary key fingerprint
92
+ + gpg-interface: refactor the free-and-xmemdupz pattern
93
+ (this branch is used by hi/gpg-use-check-signature.)
94
78
- Doc update.
95
+ The code to parse GPG output used to assume incorrectly that the
96
+ finterprint for the primary key would always be present for a valid
97
+ signature, which has been corrected.
98
99
81
-* ds/test-read-graph (2019-11-13) 1 commit
82
- (merged to 'next' on 2019-11-19 at 80152fc082)
83
- + test-tool: use 'read-graph' helper
100
+* hv/assume-priumax-is-available-anywhere (2019-11-25) 1 commit
101
+ (merged to 'next' on 2019-11-27 at caac4f633d)
102
+ + git-compat-util.h: drop the `PRIuMAX` and other fallback definitions
103
85
- Dev support for commit-graph feature.
86
-
87
-
88
-* en/doc-typofix (2019-11-10) 10 commits
89
- (merged to 'next' on 2019-11-19 at e7bfe50de5)
90
- + Fix spelling errors in no-longer-updated-from-upstream modules
91
- + multimail: fix a few simple spelling errors
92
- + sha1dc: fix trivial comment spelling error
93
- + Fix spelling errors in test commands
94
- + Fix spelling errors in messages shown to users
95
- + Fix spelling errors in names of tests
96
- + Fix spelling errors in comments of testcases
97
- + Fix spelling errors in code comments
98
- + Fix spelling errors in documentation outside of Documentation/
99
- + Documentation: fix a bunch of typos, both old and new
104
+ We have had compatibility fallback macro definitions for "PRIuMAX",
105
+ "PRIu32", etc. but did not for "PRIdMAX", while the code used the
106
+ last one apparently without any hiccup reported recently. The
107
+ fallback macro definitions for these <inttypes.h> macros that must
108
+ appear in C99 systems have been removed.
109
101
- Docfix.
102
-
103
-
104
-* en/misc-doc-fixes (2019-11-07) 3 commits
105
- (merged to 'next' on 2019-11-19 at 7222cf3720)
106
- + name-hash.c: remove duplicate word in comment
107
- + hashmap: fix documentation misuses of -> versus .
108
- + git-filter-branch.txt: correct argument name typo
110
110
- Misc doc fixes.
111
+* jc/azure-ci-osx-fix-fix (2019-11-20) 1 commit
112
+ (merged to 'next' on 2019-11-25 at 7de6356e67)
113
+ + ci(osx): update homebrew-cask repository with less noise
114
115
+ CI jobs for macOS has been made less chatty when updating perforce
116
+ package used during testing.
117
113
-* en/t6024-style (2019-11-07) 1 commit
114
- (merged to 'next' on 2019-11-19 at b41c2ed21b)
115
- + t6024: modernize style
118
117
- Test updates.
119
+* jh/userdiff-python-async (2019-11-20) 1 commit
120
+ (merged to 'next' on 2019-11-25 at 74e4a4ed9f)
121
+ + userdiff: support Python async functions
122
123
+ The userdiff machinery has been taught that "async def" is another
124
+ way to begin a "function" in Python.
125
120
-* es/myfirstcontrib-updates (2019-11-02) 3 commits
121
- (merged to 'next' on 2019-11-19 at 41d70e1db4)
122
- + myfirstcontrib: hint to find gitgitgadget allower
123
- + myfirstcontrib: add dependency installation step
124
- + myfirstcontrib: add 'psuh' to command-list.txt
126
126
- Doc updates.
127
+* jk/fail-show-toplevel-outside-working-tree (2019-11-20) 1 commit
128
+ (merged to 'next' on 2019-11-25 at cd97ae38bc)
129
+ + rev-parse: make --show-toplevel without a worktree an error
130
+
131
+ "git rev-parse --show-toplevel" run outside of any working tree did
132
+ not error out, which has been corrected.
133
+
134
135
+* js/builtin-add-i (2019-11-18) 9 commits
136
+ (merged to 'next' on 2019-11-21 at caefa55b89)
137
+ + built-in add -i: implement the `help` command
138
+ + built-in add -i: use color in the main loop
139
+ + built-in add -i: support `?` (prompt help)
140
+ + built-in add -i: show unique prefixes of the commands
141
+ + built-in add -i: implement the main loop
142
+ + built-in add -i: color the header in the `status` command
143
+ + built-in add -i: implement the `status` command
144
+ + diff: export diffstat interface
145
+ + Start to implement a built-in version of `git add --interactive`
146
+ (this branch is used by js/builtin-add-i-cmds.)
147
129
-* hw/config-doc-in-header (2019-10-24) 1 commit
130
- (merged to 'next' on 2019-11-19 at 7f314c649e)
131
- + config: move documentation to config.h
148
+ The beginning of rewriting "git add -i" in C.
149
133
- Follow recent push to move API docs from Documentation/ to header
134
- files and update config.h
150
151
+* js/git-svn-use-rebase-merges (2019-11-23) 1 commit
152
+ (merged to 'next' on 2019-11-25 at 9585fe30f8)
153
+ + git svn: stop using `rebase --preserve-merges`
154
137
-* jc/fsmonitor-sanity-fix (2019-11-13) 1 commit
138
- (merged to 'next' on 2019-11-19 at 1fceb9b55b)
139
- + fsmonitor: do not compare bitmap size with size of split index
140
- (this branch is used by us/unpack-trees-fsmonitor.)
155
+ Prepare further for "rebase --preserve-merges" deprecation.
156
142
- Remove an incorrect BUG() that should not trigger.
157
158
+* js/pkt-line-h-typofix (2019-12-02) 1 commit
159
+ (merged to 'next' on 2019-12-02 at 8f36a354ae)
160
+ + pkt-line: fix a typo
161
145
-* jc/log-graph-simplify (2019-11-13) 14 commits
146
- (merged to 'next' on 2019-11-19 at de6360c648)
147
- + t4215: use helper function to check output
148
- (merged to 'next' on 2019-11-10 at 79d2b8ff98)
149
- + graph: fix coloring of octopus dashes
150
- + graph: flatten edges that fuse with their right neighbor
151
- + graph: smooth appearance of collapsing edges on commit lines
152
- + graph: rename `new_mapping` to `old_mapping`
153
- + graph: commit and post-merge lines for left-skewed merges
154
- + graph: tidy up display of left-skewed merges
155
- + graph: example of graph output that can be simplified
156
- + graph: extract logic for moving to GRAPH_PRE_COMMIT state
157
- + graph: remove `mapping_idx` and `graph_update_width()`
158
- + graph: reduce duplication in `graph_insert_into_new_columns()`
159
- + graph: reuse `find_new_column_by_commit()`
160
- + graph: handle line padding in `graph_next_line()`
161
- + graph: automatically track display width of graph lines
162
+ Typofix.
163
163
- The implementation of "git log --graph" got refactored and then its
164
- output got simplified.
164
165
+* js/rebase-r-safer-label (2019-11-18) 2 commits
166
+ (merged to 'next' on 2019-11-21 at 791d51b0d7)
167
+ + rebase -r: let `label` generate safer labels
168
+ + rebase-merges: move labels' whitespace mangling into `label_oid()`
169
+ (this branch is used by dd/rebase-merge-reserves-onto-label.)
170
167
-* jk/cleanup-object-parsing-and-fsck (2019-10-28) 23 commits
168
- (merged to 'next' on 2019-11-10 at 50b2267a2a)
169
- + fsck: accept an oid instead of a "struct tree" for fsck_tree()
170
- + fsck: accept an oid instead of a "struct commit" for fsck_commit()
171
- + fsck: accept an oid instead of a "struct tag" for fsck_tag()
172
- + fsck: rename vague "oid" local variables
173
- + fsck: don't require an object struct in verify_headers()
174
- + fsck: don't require an object struct for fsck_ident()
175
- + fsck: drop blob struct from fsck_finish()
176
- + fsck: accept an oid instead of a "struct blob" for fsck_blob()
177
- + fsck: don't require an object struct for report()
178
- + fsck: only require an oid for skiplist functions
179
- + fsck: only provide oid/type in fsck_error callback
180
- + fsck: don't require object structs for display functions
181
- + fsck: use oids rather than objects for object_name API
182
- + fsck_describe_object(): build on our get_object_name() primitive
183
- + fsck: unify object-name code
184
- + fsck: require an actual buffer for non-blobs
185
- + fsck: stop checking tag->tagged
186
- + fsck: stop checking commit->parent counts
187
- + fsck: stop checking commit->tree value
188
- + commit, tag: don't set parsed bit for parse failures
189
- + parse_tag_buffer(): treat NULL tag pointer as parse error
190
- + parse_commit_buffer(): treat lookup_tree() failure as parse error
191
- + parse_commit_buffer(): treat lookup_commit() failure as parse error
192
-
193
- Originally merged to 'next' on 2019-10-30
194
-
195
- Crufty code and logic accumulated over time around the object
196
- parsing and low-level object access used in "git fsck" have been
197
- cleaned up.
171
+ A label used in the todo list that are generated by "git rebase
172
+ --rebase-merges" is used as a part of a refname; the logic to come
173
+ up with the label has been tightened to avoid names that cannot be
174
+ used as such.
175
176
200
-* jk/gitweb-anti-xss (2019-11-18) 4 commits
201
- (merged to 'next' on 2019-11-19 at 69ef04c013)
202
- + gitweb: escape URLs generated by href()
203
- + t/gitweb-lib.sh: set $REQUEST_URI
204
- + t/gitweb-lib.sh: drop confusing quotes
205
- + t9502: pass along all arguments in xss helper
206
-
207
- Some codepaths in "gitweb" that forgot to escape URLs generated
208
- based on end-user input have been corrected.
177
+* mg/doc-submodule-status-cached (2019-11-24) 1 commit
178
+ (merged to 'next' on 2019-11-25 at fa833392a3)
179
+ + doc: document 'git submodule status --cached'
180
181
+ "git submodule status" and "git submodule status --cached" show
182
+ different things, but the documentation did not cover them
183
+ correctly, which has been corrected.
184
211
-* jk/optim-in-pack-idx-conversion (2019-11-12) 1 commit
212
- (merged to 'next' on 2019-11-19 at a81ebc7b8d)
213
- + pack-objects: avoid pointless oe_map_new_pack() calls
214
-
215
- Code clean-up.
185
186
+* mg/submodule-status-from-a-subdirectory (2019-11-25) 1 commit
187
+ (merged to 'next' on 2019-11-27 at 24cb2d79ff)
188
+ + submodule: fix 'submodule status' when called from a subdirectory
189
218
-* jk/remove-sha1-to-hex (2019-11-13) 2 commits
219
- (merged to 'next' on 2019-11-19 at fb68824870)
220
- + hex: drop sha1_to_hex()
221
- + hex: drop sha1_to_hex_r()
190
+ "git submodule status" that is run from a subdirectory of the
191
+ superproject did not work well, which has been corrected.
192
223
- Code clean-up.
193
194
+* mh/clear-topo-walk-upon-reset (2019-11-25) 2 commits
195
+ (merged to 'next' on 2019-11-27 at 15871c1c2a)
196
+ + revision: free topo_walk_info before creating a new one in init_topo_walk
197
+ + revision: clear the topo-walk flags in reset_revision_walk
198
226
-* jk/send-pack-remote-failure (2019-11-13) 1 commit
227
- (merged to 'next' on 2019-11-19 at d02c0107ee)
228
- + send-pack: check remote ref status on pack-objects failure
199
+ The revision walking machinery uses resources like per-object flag
200
+ bits that need to be reset before a new iteration of walking
201
+ begins, but the resources related to topological walk were not
202
+ cleared correctly, which has been corrected.
203
230
- Error handling after "git push" finishes sending the packdata and
231
- waits for the response to the remote side has been improved.
204
205
+* nl/reset-patch-takes-a-tree (2019-11-25) 1 commit
206
+ (merged to 'next' on 2019-11-25 at 34dc348eaf)
207
+ + reset: parse rev as tree-ish in patch mode
208
234
-* js/complete-svn-recursive (2019-11-13) 1 commit
235
- (merged to 'next' on 2019-11-19 at fa2a97911c)
236
- + completion: tab-complete "git svn --recursive"
209
+ "git reset --patch $object" without any pathspec should allow a
210
+ tree object to be given, but incorrectly required a committish,
211
+ which has been corrected.
212
238
- The completion script (in contrib/) has been taught that "git svn"
239
- supports the "--recursive" option.
213
214
+* pb/submodule-update-fetches (2019-11-24) 1 commit
215
+ (merged to 'next' on 2019-11-25 at 7736f98351)
216
+ + doc: mention that 'git submodule update' fetches missing commits
217
242
-* js/fetch-multi-lockfix (2019-11-06) 2 commits
243
- (merged to 'next' on 2019-11-19 at b9683b7ec9)
244
- + fetch: avoid locking issues between fetch.jobs/fetch.writeCommitGraph
245
- + fetch: add the command-line option `--write-commit-graph`
218
+ Doc update.
219
247
- Fetching from multiple remotes into the same repository in parallel
248
- had a bad interaction with the recent change to (optionally) update
249
- the commit-graph after a fetch job finishes, as these parallel
250
- fetches compete with each other. Which has been corrected.
220
221
+* pw/sequencer-compare-with-right-parent-to-check-empty-commits (2019-11-23) 1 commit
222
+ (merged to 'next' on 2019-11-25 at a4e4f06286)
223
+ + sequencer: fix empty commit check when amending
224
253
-* js/git-path-head-dot-lock-fix (2019-10-29) 2 commits
254
- (merged to 'next' on 2019-11-19 at b406944979)
255
- + git_path(): handle `.lock` files correctly
256
- + t1400: wrap setup code in test case
225
+ The sequencer machinery compared the HEAD and the state it is
226
+ attempting to commit to decide if the result would be a no-op
227
+ commit, even when amending a commit, which was incorrect, and
228
+ has been corrected.
229
258
- "git rev-parse --git-path HEAD.lock" did not give the right path
259
- when run in a secondary worktree.
230
231
+* sg/test-bool-env (2019-11-23) 2 commits
232
+ (merged to 'next' on 2019-11-27 at e1874e422f)
233
+ + t5608-clone-2gb.sh: turn GIT_TEST_CLONE_2GB into a bool
234
+ + tests: add 'test_bool_env' to catch non-bool GIT_TEST_* values
235
262
-* js/vreportf-wo-buffering (2019-11-02) 1 commit
263
- (merged to 'next' on 2019-11-19 at cae61b9899)
264
- + vreportf(): avoid relying on stdio buffering
236
+ Recently we have declared that GIT_TEST_* variables take the
237
+ usual boolean values (it used to be that some used "non-empty
238
+ means true" and taking GIT_TEST_VAR=YesPlease as true); make
239
+ sure we notice and fail when non-bool strings are given to
240
+ these variables.
241
266
- Messages from die() etc. can be mixed up from multiple processes
267
- without even line buffering on Windows, which has been worked
268
- around.
242
243
+* sg/unpack-progress-throughput (2019-11-20) 1 commit
244
+ (merged to 'next' on 2019-11-25 at 0418db6a27)
245
+ + builtin/unpack-objects.c: show throughput progress
246
271
-* jt/fetch-remove-lazy-fetch-plugging (2019-11-13) 3 commits
272
- (merged to 'next' on 2019-11-19 at 1b26c28dcd)
273
- + promisor-remote: remove fetch_if_missing=0
274
- + clone: remove fetch_if_missing=0
275
- + fetch: remove fetch_if_missing=0
247
+ "git unpack-objects" used to show progress based only on the number
248
+ of received and unpacked objects, which stalled when it has to
249
+ handle an unusually large object. It now shows the throughput as
250
+ well.
251
277
- "git fetch" codepath had a big "do not lazily fetch missing objects
278
- when I ask if something exists" switch. This has been corrected by
279
- marking the "does this thing exist?" calls with "if not please do not
280
- lazily fetch it" flag.
252
253
+* us/unpack-trees-fsmonitor (2019-11-21) 1 commit
254
+ (merged to 'next' on 2019-11-27 at 4c06f74957)
255
+ + unpack-trees: skip stat on fsmonitor-valid files
256
283
-* kw/fsmonitor-watchman-fix (2019-11-06) 1 commit
284
- (merged to 'next' on 2019-11-19 at ee786a5c39)
285
- + fsmonitor: fix watchman integration
257
+ Users of oneway_merge() (like "reset --hard") learned to take
258
+ advantage of fsmonitor to avoid unnecessary lstat(2) calls.
259
287
- The watchman integration for fsmonitor was racy, which has been
288
- corrected to be more conservative.
260
+--------------------------------------------------
261
+[New Topics]
262
263
+* am/pathspec-f-f-checkout (2019-12-04) 7 commits
264
+ - checkout, restore: support the --pathspec-from-file option
265
+ - doc: restore: synchronize <pathspec> description
266
+ - doc: checkout: synchronize <pathspec> description
267
+ - doc: checkout: fix broken text reference
268
+ - doc: checkout: remove duplicate synopsis
269
+ - add: support the --pathspec-from-file option
270
+ - cmd_add: prepare for next patch
271
+ (this branch uses am/pathspec-from-file; is tangled with es/pathspec-f-f-grep.)
272
+
273
+ A few more commands learned the "--pathspec-from-file" command line
274
+ option.
275
291
-* ln/userdiff-elixir (2019-11-10) 1 commit
292
- (merged to 'next' on 2019-11-19 at 6318918ba8)
293
- + userdiff: add Elixir to supported userdiff languages
276
+ Will merge to 'next'.
277
295
- The patterns to detect function boundary for Elixir language has
296
- been added.
278
279
+* dl/lore-is-the-archive (2019-12-04) 3 commits
280
+ (merged to 'next' on 2019-12-05 at c87e168cd6)
281
+ + doc: replace LKML link with lore.kernel.org
282
+ + RelNotes: replace Gmane with real Message-IDs
283
+ + doc: replace MARC links with lore.kernel.org
284
+ (this branch uses jk/lore-is-the-archive.)
285
299
-* ma/bisect-doc-sample-update (2019-10-29) 1 commit
300
- (merged to 'next' on 2019-11-19 at bde6adce1b)
301
- + Documentation/git-bisect.txt: add --no-ff to merge command
286
+ Publicize lore.kernel.org mailing list archive and use URLs
287
+ pointing into it to refer to notable messages in the documentation.
288
303
- "git merge --no-commit" needs "--no-ff" if you do not want to move
304
- HEAD, which has been corrected in the manual page for "git bisect".
289
+ Will merge to 'master'.
290
291
307
-* ma/t7004 (2019-11-14) 1 commit
308
- (merged to 'next' on 2019-11-19 at 789a0a0f05)
309
- + t7004: check existence of correct tag
292
+* dl/rebase-with-autobase (2019-12-05) 5 commits
293
+ - rebase: fix format.useAutoBase breakage
294
+ - format-patch: teach --no-base
295
+ - t4014: use test_config()
296
+ - format-patch: fix indentation
297
+ - t3400: demonstrate failure with format.useAutoBase
298
311
- Test fix.
299
+ "git rebase" did not work well when format.useAutoBase
300
+ configuration variable is set, which has been corrected.
301
302
+ Will merge to 'next'.
303
314
-* mr/clone-dir-exists-to-path-exists (2019-10-29) 1 commit
315
- (merged to 'next' on 2019-11-19 at 07b6ab25de)
316
- + clone: rename static function `dir_exists()`.
304
318
- Code cleanup.
305
+* es/pathspec-f-f-grep (2019-12-04) 1 commit
306
+ - grep: support the --pathspec-from-file option
307
+ (this branch uses am/pathspec-from-file; is tangled with am/pathspec-f-f-checkout.)
308
309
+ "git grep" learned the "--pathspec-from-file" command line
310
+ option.
311
321
-* nn/doc-rebase-merges (2019-11-12) 1 commit
322
- (merged to 'next' on 2019-11-19 at db74ff488c)
323
- + doc: improve readability of --rebase-merges in git-rebase
312
+ Waiting for review response.
313
325
- Doc update.
314
315
+* rs/xdiff-ignore-ws-w-func-context (2019-12-05) 1 commit
316
+ - xdiff: unignore changes in function context
317
328
-* ns/test-desc-typofix (2019-11-07) 1 commit
329
- (merged to 'next' on 2019-11-19 at 804861f385)
330
- + t: fix typo in test descriptions
318
+ The "diff" machinery learned not to lose added/removed blank lines
319
+ in the context when --ignore-blank-lines and --function-context are
320
+ used at the same time.
321
332
- Typofix.
322
+ Will merge to 'next'.
323
324
335
-* pb/help-list-gitsubmodules-among-guides (2019-10-29) 1 commit
336
- (merged to 'next' on 2019-11-19 at 3c6f5c1901)
337
- + help: add gitsubmodules to the list of guides
325
+* ag/edit-todo-drop-check (2019-12-04) 2 commits
326
+ - rebase-interactive: warn if commit is dropped with `rebase --edit-todo'
327
+ - sequencer: move check_todo_list_from_file() to rebase-interactive.c
328
+ (this branch uses ag/sequencer-todo-updates.)
329
339
- Help update.
330
+ Allow the rebase.missingCommitsCheck configuration to kick in when
331
+ "rebase --edit-todo" and "rebase --continue" restarts the procedure.
332
333
+ Seems to break when merged to 'pu'.
334
342
-* pb/no-recursive-reset-hard-in-worktree-add (2019-10-30) 1 commit
343
- (merged to 'next' on 2019-11-19 at cdfb0643e9)
344
- + worktree: teach "add" to ignore submodule.recurse config
335
346
- "git worktree add" internally calls "reset --hard" that should not
347
- descend into submodules, even when submodule.recurse configuration
348
- is set, but it was affected. This has been corrected.
336
+--------------------------------------------------
337
+[Stalled]
338
339
+* mt/threaded-grep-in-object-store (2019-10-02) 11 commits
340
+ - grep: move driver pre-load out of critical section
341
+ - grep: re-enable threads in non-worktree case
342
+ - grep: protect packed_git [re-]initialization
343
+ - grep: allow submodule functions to run in parallel
344
+ - submodule-config: add skip_if_read option to repo_read_gitmodules()
345
+ - grep: replace grep_read_mutex by internal obj read lock
346
+ - object-store: allow threaded access to object reading
347
+ - replace-object: make replace operations thread-safe
348
+ - grep: fix racy calls in grep_objects()
349
+ - grep: fix race conditions at grep_submodule()
350
+ - grep: fix race conditions on userdiff calls
351
351
-* py/shortlog-list-options-for-log (2019-11-10) 1 commit
352
- (merged to 'next' on 2019-11-19 at 92a10e39f7)
353
- + git-shortlog.txt: include commit limiting options
352
+ Traditionally, we avoided threaded grep while searching in objects
353
+ (as opposed to files in the working tree) as accesses to the object
354
+ layer is not thread-safe. This limitation is getting lifted.
355
355
- Documentation pages for "git shortlog" now lists commit limiting
356
- options explicitly.
356
+ Expecting a reroll.
357
+ cf. <CAHd-oW7UPSSExyLtfLMCObWogKrBOctYabrFrOdf9-4Q2PZmMg@mail.gmail.com>
358
359
359
-* rj/bundle-ui-updates (2019-11-11) 3 commits
360
- (merged to 'next' on 2019-11-19 at d88f626256)
361
- + bundle-verify: add --quiet
362
- + bundle-create: progress output control
363
- + bundle: framework for options before bundle file
360
+* po/bundle-doc-clonable (2019-10-21) 1 commit
361
+ (merged to 'next' on 2019-12-02 at bd7fa24ad6)
362
+ + Doc: Bundle file usage
363
365
- "git bundle" has been taught to use the parse options API. "git
366
- bundle verify" learned "--quiet" and "git bundle create" learned
367
- options to control the progress output.
364
+ Doc update.
365
366
+ Will merge to 'master'.
367
370
-* rs/parse-options-dup-null-fix (2019-11-13) 1 commit
371
- (merged to 'next' on 2019-11-19 at 73040e1605)
372
- + parse-options: avoid arithmetic on pointer that's potentially NULL
368
374
- Code cleanup.
369
+* js/advise-rebase-skip (2019-10-23) 3 commits
370
+ - commit: give correct advice for empty commit during a rebase
371
+ - sequencer: export the function to get the path of `.git/rebase-merge/`
372
+ - cherry-pick: add test for `--skip` advice in `git commit`
373
374
+ The logic used in "git commit" to give hints and errors depending
375
+ on what operation was in progress learned to distinguish rebase and
376
+ cherry-pick better.
377
377
-* rs/skip-iprefix (2019-11-10) 2 commits
378
- (merged to 'next' on 2019-11-19 at c87ae31b14)
379
- + convert: use skip_iprefix() in validate_encoding()
380
- + utf8: use skip_iprefix() in same_utf_encoding()
378
+ Kicked back to 'pu'.
379
+ cf. <nycvar.QRO.7.76.6.1911081508560.46@tvgsbejvaqbjf.bet>
380
382
- Code simplification.
381
382
+* vn/reset-deleted-ita (2019-07-26) 1 commit
383
+ - reset: unstage empty deleted ita files
384
385
-* rs/trace2-dots (2019-11-06) 1 commit
386
- (merged to 'next' on 2019-11-19 at a045233ad0)
387
- + trace2: add dots directly to strbuf in perf_fmt_prepare()
385
+ "git reset HEAD [<pathspec>]" did not reset an empty file that was
386
+ added with the intent-to-add bit.
387
389
- Code cleanup.
388
+ Expecting a reroll.
389
390
392
-* rs/use-copy-array-in-mingw-shell-command-preparation (2019-11-13) 1 commit
393
- (merged to 'next' on 2019-11-19 at 52ee6a8887)
394
- + mingw: use COPY_ARRAY for copying array
391
+* jn/unknown-index-extensions (2018-11-21) 2 commits
392
+ - index: offer advice for unknown index extensions
393
+ - index: do not warn about unrecognized extensions
394
396
- Code cleanup.
395
+ A bit too alarming warning given when unknown index extensions
396
+ exist is getting revamped.
397
398
+ Expecting a reroll.
399
399
-* rt/fetch-message-fix (2019-11-02) 1 commit
400
- (merged to 'next' on 2019-11-19 at 69eb337cd8)
401
- + fetch.c: fix typo in a warning message
400
403
- A small message update.
401
+* jc/format-patch-delay-message-id (2019-04-05) 1 commit
402
+ - format-patch: move message-id and related headers to the end
403
404
+ The location "git format-patch --thread" adds the Message-Id:
405
+ header in the series of header fields has been moved down, which
406
+ may help working around a suspected bug in GMail MSA, reported at
407
+ <CAHk-=whP1stFZNAaJiMi5eZ9rj0MRt20Y_yHVczZPH+O01d+sA@mail.gmail.com>
408
406
-* sg/blame-indent-heuristics-is-now-the-default (2019-10-29) 1 commit
407
- (merged to 'next' on 2019-11-19 at 68e60f08cc)
408
- + builtin/blame.c: remove '--indent-heuristic' from usage string
409
+ Waiting for feedback to see if it truly helps.
410
+ Needs tests.
411
410
- Message update.
412
413
+* js/protocol-advertise-multi (2018-12-28) 1 commit
414
+ - protocol: advertise multiple supported versions
415
413
-* sg/skip-skipped-prereq (2019-11-13) 1 commit
414
- (merged to 'next' on 2019-11-19 at 10d344fed7)
415
- + test-lib: don't check prereqs of test cases that won't be run anyway
416
+ The transport layer has been updated so that the protocol version
417
+ used can be negotiated between the parties, by the initiator
418
+ listing the protocol versions it is willing to talk, and the other
419
+ side choosing from one of them.
420
417
- Test update to avoid wasted cycles.
421
+ Expecting a reroll.
422
+ cf. <CANq=j3u-zdb_FvNJGPCmygNMScseav63GhVvBX3NcVS4f7TejA@mail.gmail.com>
423
424
420
-* tg/stash-refresh-index (2019-11-14) 1 commit
421
- (merged to 'next' on 2019-11-19 at 4d67af8d3f)
422
- + stash: make sure we have a valid index before writing it
425
+* mk/use-size-t-in-zlib (2018-10-15) 1 commit
426
+ - zlib.c: use size_t for size
427
424
- Recent update to "git stash pop" made the command empty the index
425
- when run with the "--quiet" option, which has been corrected.
428
+ The wrapper to call into zlib followed our long tradition to use
429
+ "unsigned long" for sizes of regions in memory, which have been
430
+ updated to use "size_t".
431
432
--------------------------------------------------
428
-[New Topics]
433
+[Cooking]
434
435
* ds/commit-graph-delay-gen-progress (2019-11-27) 2 commits
431
- - commit-graph: use start_delayed_progress()
432
- - progress: create GIT_PROGRESS_DELAY
436
+ (merged to 'next' on 2019-12-05 at 9252a4b1b6)
437
+ + commit-graph: use start_delayed_progress()
438
+ + progress: create GIT_PROGRESS_DELAY
439
440
One kind of progress messages were always given during commit-graph
441
generation, instead of following the "if it takes more than two
442
seconds, show progress" pattern, which has been corrected.
443
438
- Will merge to 'next'.
444
+ Will merge to 'master'.
445
446
447
* ds/sparse-cone (2019-11-22) 19 commits
@@ -462,71 +468,11 @@ of the repositories listed at
468
Management of sparsely checked-out working tree has gained a
469
dedicated "sparse-checkout" command.
470
465
- How well is this topic done at this point?
466
-
467
-
468
-* dl/range-diff-with-notes (2019-11-21) 10 commits
469
- (merged to 'next' on 2019-11-25 at dd6dad97c1)
470
- + format-patch: pass notes configuration to range-diff
471
- + range-diff: pass through --notes to `git log`
472
- + range-diff: output `## Notes ##` header
473
- + t3206: range-diff compares logs with commit notes
474
- + t3206: s/expected/expect/
475
- + t3206: disable parameter substitution in heredoc
476
- + t3206: remove spaces after redirect operators
477
- + pretty-options.txt: --notes accepts a ref instead of treeish
478
- + rev-list-options.txt: remove reference to --show-notes
479
- + argv-array: add space after `while`
480
-
481
- "git range-diff" learned to take the "--notes=<ref>" and the
482
- "--no-notes" options to control the commit notes included in the
483
- log message that gets compared.
484
-
485
- Will merge to 'master'.
486
-
471
+ Possibly wait for Szeder?
472
+ cf. <bea26b66-ac52-8f0e-9557-c958db7be628@gmail.com>
473
488
-* jc/azure-ci-osx-fix-fix (2019-11-20) 1 commit
489
- (merged to 'next' on 2019-11-25 at 7de6356e67)
490
- + ci(osx): update homebrew-cask repository with less noise
491
-
492
- CI jobs for macOS has been made less chatty when updating perforce
493
- package used during testing.
494
-
495
- Will merge to 'master'.
496
-
497
-
498
-* jk/fail-show-toplevel-outside-working-tree (2019-11-20) 1 commit
499
- (merged to 'next' on 2019-11-25 at cd97ae38bc)
500
- + rev-parse: make --show-toplevel without a worktree an error
501
-
502
- "git rev-parse --show-toplevel" run outside of any working tree did
503
- not error out, which has been corrected.
474
505
- Will merge to 'master'.
506
-
507
-
508
-* pb/submodule-update-fetches (2019-11-24) 1 commit
509
- (merged to 'next' on 2019-11-25 at 7736f98351)
510
- + doc: mention that 'git submodule update' fetches missing commits
511
-
512
- Doc update.
513
-
514
- Will merge to 'master'.
515
-
516
-
517
-* sg/unpack-progress-throughput (2019-11-20) 1 commit
518
- (merged to 'next' on 2019-11-25 at 0418db6a27)
519
- + builtin/unpack-objects.c: show throughput progress
520
-
521
- "git unpack-objects" used to show progress based only on the number
522
- of received and unpacked objects, which stalled when it has to
523
- handle an unusually large object. It now shows the throughput as
524
- well.
525
-
526
- Will merge to 'master'.
527
-
528
-
529
-* dl/test-cleanup (2019-11-29) 26 commits
475
+* dl/test-cleanup (2019-12-04) 26 commits
476
- t7700: stop losing return codes of git commands
477
- t7700: make references to SHA-1 generic
478
- t7700: replace egrep with grep
@@ -560,21 +506,12 @@ of the repositories listed at
506
507
508
* tg/perf-remove-stale-result (2019-11-27) 1 commit
563
- - perf-lib: use a single filename for all measurement types
509
+ (merged to 'next' on 2019-12-05 at 466f07fb63)
510
+ + perf-lib: use a single filename for all measurement types
511
512
PerfTest fix to avoid stale result mixed up with the latest round
513
of test results.
514
568
- Will merge to 'next'.
569
-
570
-
571
-* us/unpack-trees-fsmonitor (2019-11-21) 1 commit
572
- (merged to 'next' on 2019-11-27 at 4c06f74957)
573
- + unpack-trees: skip stat on fsmonitor-valid files
574
-
575
- Users of oneway_merge() (like "reset --hard") learned to take
576
- advantage of fsmonitor to avoid unnecessary lstat(2) calls.
577
-
515
Will merge to 'master'.
516
517
@@ -593,114 +530,24 @@ of the repositories listed at
530
531
Updates around testing fsmoitor integration.
532
596
- Will merge to 'next'.
597
-
598
-
599
-* hi/gpg-optional-pkfp-fix (2019-11-23) 2 commits
600
- (merged to 'next' on 2019-11-25 at b0d668176f)
601
- + gpg-interface: limit search for primary key fingerprint
602
- + gpg-interface: refactor the free-and-xmemdupz pattern
603
- (this branch is used by hi/gpg-use-check-signature.)
604
-
605
- The code to parse GPG output used to assume incorrectly that the
606
- finterprint for the primary key would always be present for a valid
607
- signature, which has been corrected.
608
-
609
- Will merge to 'master'.
610
-
611
-
612
-* hv/assume-priumax-is-available-anywhere (2019-11-25) 1 commit
613
- (merged to 'next' on 2019-11-27 at caac4f633d)
614
- + git-compat-util.h: drop the `PRIuMAX` and other fallback definitions
615
-
616
- We have had compatibility fallback macro definitions for "PRIuMAX",
617
- "PRIu32", etc. but did not for "PRIdMAX", while the code used the
618
- last one apparently without any hiccup reported recently. The
619
- fallback macro definitions for these <inttypes.h> macros that must
620
- appear in C99 systems have been removed.
621
-
622
- Will merge to 'master'.
623
-
624
-
625
-* js/git-svn-use-rebase-merges (2019-11-23) 1 commit
626
- (merged to 'next' on 2019-11-25 at 9585fe30f8)
627
- + git svn: stop using `rebase --preserve-merges`
628
-
629
- Prepare further for "rebase --preserve-merges" deprecation.
630
-
631
- Will merge to 'master'.
533
+ On Hold.
534
+ cf. <bea26b66-ac52-8f0e-9557-c958db7be628@gmail.com>
535
536
537
* js/mingw-inherit-only-std-handles (2019-12-02) 6 commits
635
- - mingw: translate ERROR_SUCCESS to errno = 0
636
- (merged to 'next' on 2019-12-02 at 798a952eca)
637
- + mingw: do set `errno` correctly when trying to restrict handle inheritance
638
- (merged to 'next' on 2019-11-27 at 4736894af6)
639
- + mingw: restrict file handle inheritance only on Windows 7 and later
640
- + mingw: spawned processes need to inherit only standard handles
641
- + mingw: work around incorrect standard handles
642
- + mingw: demonstrate that all file handles are inherited by child processes
643
-
644
- Work around a issue where a FD that is left open when spawning a
645
- child process and is kept open in the child can interfere with the
646
- operation in the parent process on Windows.
647
-
648
- Will wait for a bit before merging to 'next'.
649
-
650
-
651
-* mg/doc-submodule-status-cached (2019-11-24) 1 commit
652
- (merged to 'next' on 2019-11-25 at fa833392a3)
653
- + doc: document 'git submodule status --cached'
654
-
655
- "git submodule status" and "git submodule status --cached" show
656
- different things, but the documentation did not cover them
657
- correctly, which has been corrected.
658
-
659
- Will merge to 'master'.
660
-
661
-
662
-* mg/submodule-status-from-a-subdirectory (2019-11-25) 1 commit
663
- (merged to 'next' on 2019-11-27 at 24cb2d79ff)
664
- + submodule: fix 'submodule status' when called from a subdirectory
665
-
666
- "git submodule status" that is run from a subdirectory of the
667
- superproject did not work well, which has been corrected.
668
-
669
- Will merge to 'master'.
670
-
671
-
672
-* mh/clear-topo-walk-upon-reset (2019-11-25) 2 commits
673
- (merged to 'next' on 2019-11-27 at 15871c1c2a)
674
- + revision: free topo_walk_info before creating a new one in init_topo_walk
675
- + revision: clear the topo-walk flags in reset_revision_walk
676
-
677
- The revision walking machinery uses resources like per-object flag
678
- bits that need to be reset before a new iteration of walking
679
- begins, but the resources related to topological walk were not
680
- cleared correctly, which has been corrected.
681
-
682
- Will merge to 'master'.
683
-
684
-
685
-* nl/reset-patch-takes-a-tree (2019-11-25) 1 commit
686
- (merged to 'next' on 2019-11-25 at 34dc348eaf)
687
- + reset: parse rev as tree-ish in patch mode
688
-
689
- "git reset --patch $object" without any pathspec should allow a
690
- tree object to be given, but incorrectly required a committish,
691
- which has been corrected.
692
-
693
- Will merge to 'master'.
694
-
695
-
696
-* pw/sequencer-compare-with-right-parent-to-check-empty-commits (2019-11-23) 1 commit
697
- (merged to 'next' on 2019-11-25 at a4e4f06286)
698
- + sequencer: fix empty commit check when amending
538
+ (merged to 'next' on 2019-12-05 at 289f13054b)
539
+ + mingw: forbid translating ERROR_SUCCESS to an errno value
540
+ (merged to 'next' on 2019-12-02 at 798a952eca)
541
+ + mingw: do set `errno` correctly when trying to restrict handle inheritance
542
+ (merged to 'next' on 2019-11-27 at 4736894af6)
543
+ + mingw: restrict file handle inheritance only on Windows 7 and later
544
+ + mingw: spawned processes need to inherit only standard handles
545
+ + mingw: work around incorrect standard handles
546
+ + mingw: demonstrate that all file handles are inherited by child processes
547
700
- The sequencer machinery compared the HEAD and the state it is
701
- attempting to commit to decide if the result would be a no-op
702
- commit, even when amending a commit, which was incorrect, and
703
- has been corrected.
548
+ Work around a issue where a FD that is left open when spawning a
549
+ child process and is kept open in the child can interfere with the
550
+ operation in the parent process on Windows.
551
552
Will merge to 'master'.
553
@@ -717,47 +564,36 @@ of the repositories listed at
564
Will merge to 'master'.
565
566
720
-* sg/test-bool-env (2019-11-23) 2 commits
721
- (merged to 'next' on 2019-11-27 at e1874e422f)
722
- + t5608-clone-2gb.sh: turn GIT_TEST_CLONE_2GB into a bool
723
- + tests: add 'test_bool_env' to catch non-bool GIT_TEST_* values
724
-
725
- Recently we have declared that GIT_TEST_* variables take the
726
- usual boolean values (it used to be that some used "non-empty
727
- means true" and taking GIT_TEST_VAR=YesPlease as true); make
728
- sure we notice and fail when non-bool strings are given to
729
- these variables.
730
-
731
- Will merge to 'master'.
732
-
733
-
567
* as/t7812-missing-redirects-fix (2019-12-01) 2 commits
735
- - t7812: expect failure for grep -i with invalid UTF-8 data
736
- - t7812: add missing redirects
568
+ (merged to 'next' on 2019-12-05 at aa89a7945f)
569
+ + t7812: expect failure for grep -i with invalid UTF-8 data
570
+ + t7812: add missing redirects
571
572
Test fix.
573
740
- Will merge to 'next'.
574
+ Will merge to 'master'.
575
576
577
* jk/perf-wo-git-dot-pm (2019-11-27) 1 commit
744
- - t/perf: don't depend on Git.pm
578
+ (merged to 'next' on 2019-12-05 at 6d9e3930ec)
579
+ + t/perf: don't depend on Git.pm
580
581
Test cleanup.
582
748
- Will merge to 'next'.
583
+ Will merge to 'master'.
584
585
751
-* jt/clone-recursesub-ref-advise (2019-11-27) 2 commits
752
- - submodule--helper: advise on fatal alternate error
753
- - Documentation: explain submodule.alternateErrorStrategy
586
+* jt/clone-recursesub-ref-advise (2019-12-03) 2 commits
587
+ (merged to 'next' on 2019-12-05 at fc071dddfe)
588
+ + submodule--helper: advise on fatal alternate error
589
+ + Doc: explain submodule.alternateErrorStrategy
590
591
The interaction between "git clone --recurse-submodules" and
592
alternate object store was ill-designed. The documentation and
593
code have been taught to make more clear recommendations when the
594
users see failures.
595
760
- Will merge to 'next'.
596
+ Will merge to 'master'.
597
598
599
* ra/t5150-depends-on-perl (2019-11-27) 1 commit
@@ -770,242 +606,171 @@ of the repositories listed at
606
607
608
* rs/simplify-prepare-cmd (2019-11-27) 1 commit
773
- - run-command: use prepare_git_cmd() in prepare_cmd()
609
+ (merged to 'next' on 2019-12-02 at 12e1a9e935)
610
+ + run-command: use prepare_git_cmd() in prepare_cmd()
611
612
Code cleanup.
613
777
- Will merge to 'next'.
614
+ Will merge to 'master'.
615
616
617
* rs/test-cleanup (2019-11-30) 7 commits
781
- - t7811: don't create unused file
782
- - t9300: don't create unused file
783
- - test: use test_must_be_empty F instead of test_cmp empty F
784
- - test: use test_must_be_empty F instead of test -z $(cat F)
785
- - t1400: use test_must_be_empty
786
- - t1410: use test_line_count
787
- - t1512: use test_line_count
618
+ (merged to 'next' on 2019-12-02 at eabc2e30f5)
619
+ + t7811: don't create unused file
620
+ + t9300: don't create unused file
621
+ + test: use test_must_be_empty F instead of test_cmp empty F
622
+ + test: use test_must_be_empty F instead of test -z $(cat F)
623
+ + t1400: use test_must_be_empty
624
+ + t1410: use test_line_count
625
+ + t1512: use test_line_count
626
627
Test cleanup.
628
791
- Will merge to 'next'.
629
+ Will merge to 'master'.
630
631
632
* rs/use-skip-prefix-more (2019-11-27) 5 commits
795
- - name-rev: use skip_prefix() instead of starts_with()
796
- - push: use skip_prefix() instead of starts_with()
797
- - shell: use skip_prefix() instead of starts_with()
798
- - fmt-merge-msg: use skip_prefix() instead of starts_with()
799
- - fetch: use skip_prefix() instead of starts_with()
633
+ (merged to 'next' on 2019-12-02 at a2124db23e)
634
+ + name-rev: use skip_prefix() instead of starts_with()
635
+ + push: use skip_prefix() instead of starts_with()
636
+ + shell: use skip_prefix() instead of starts_with()
637
+ + fmt-merge-msg: use skip_prefix() instead of starts_with()
638
+ + fetch: use skip_prefix() instead of starts_with()
639
640
Code cleanup.
641
803
- Will merge to 'next'.
642
+ Will merge to 'master'.
643
644
645
* sg/test-squelch-noise-in-commit-bulk (2019-11-27) 1 commit
807
- - test-lib-functions: suppress a 'git rev-parse' error in 'test_commit_bulk'
646
+ (merged to 'next' on 2019-12-05 at 7819efee07)
647
+ + test-lib-functions: suppress a 'git rev-parse' error in 'test_commit_bulk'
648
649
Code cleanup.
650
811
- Will merge to 'next'.
651
+ Will merge to 'master'.
652
653
654
* ag/sequencer-continue-leakfix (2019-11-30) 1 commit
815
- - sequencer: fix a memory leak in sequencer_continue()
655
+ (merged to 'next' on 2019-12-05 at 923a6a7ab6)
656
+ + sequencer: fix a memory leak in sequencer_continue()
657
658
Leakfix.
659
819
- Will merge to 'next'.
660
+ Will merge to 'master'.
661
662
663
* bc/t9001-zsh-in-posix-emulation-mode (2019-11-30) 1 commit
823
- - t9001: avoid including non-trailing NUL bytes in variables
664
+ (merged to 'next' on 2019-12-05 at d0ce4bcc90)
665
+ + t9001: avoid including non-trailing NUL bytes in variables
666
667
Test portability fix.
668
827
- Will merge to 'next'.
669
+ Will merge to 'master'.
670
671
830
-* cs/store-packfiles-in-hashmap (2019-12-01) 1 commit
831
- - packfile.c: speed up loading lots of packfiles
672
+* cs/store-packfiles-in-hashmap (2019-12-03) 1 commit
673
+ (merged to 'next' on 2019-12-05 at 887d5559bb)
674
+ + packfile.c: speed up loading lots of packfiles
675
676
In a repository with many packfiles, the cost of the procedure that
677
avoids registering the same packfile twice was unnecessarily high
678
by using an inefficient search algorithm, which has been corrected.
679
837
- Will merge to 'next'.
680
+ Will merge to 'master'.
681
682
683
* dd/time-reentrancy (2019-12-01) 3 commits
841
- - mingw: use {gm,local}time_s as backend for {gm,local}time_r
842
- - archive-zip.c: switch to reentrant localtime_r
843
- - date.c: switch to reentrant {gm,local}time_r
684
+ (merged to 'next' on 2019-12-05 at d0c91f4d1b)
685
+ + mingw: use {gm,local}time_s as backend for {gm,local}time_r
686
+ + archive-zip.c: switch to reentrant localtime_r
687
+ + date.c: switch to reentrant {gm,local}time_r
688
689
Avoid gmtime() and localtime() and prefer their reentrant
690
counterparts.
691
848
- Will merge to 'next'.
692
+ Will merge to 'master'.
693
694
695
* em/test-skip-regex-illseq (2019-11-30) 1 commit
852
- - t4210: skip i18n tests that don't work on FreeBSD
696
+ (merged to 'next' on 2019-12-05 at 85c7940aeb)
697
+ + t4210: skip i18n tests that don't work on FreeBSD
698
699
Test portability fix.
700
856
- Will merge to 'next'.
701
+ Will merge to 'master'.
702
703
704
* hi/gpg-use-check-signature (2019-11-30) 1 commit
860
- - gpg-interface: prefer check_signature() for GPG verification
861
- (this branch uses hi/gpg-optional-pkfp-fix.)
705
+ (merged to 'next' on 2019-12-05 at 35fc8e839a)
706
+ + gpg-interface: prefer check_signature() for GPG verification
707
708
Hide lower-level verify_signed-buffer() API as a pure helper to
709
implement the public check_signature() function, in order to
710
encourage new callers to use the correct and more strict
711
validation.
712
868
- Will merge to 'next'.
713
+ Will merge to 'master'.
714
715
716
* hi/grep-do-not-return-void (2019-11-30) 1 commit
872
- - grep: don't return an expression from pcre2_free()
717
+ (merged to 'next' on 2019-12-02 at 4409c73cfa)
718
+ + grep: don't return an expression from pcre2_free()
719
720
Code cleanup.
721
876
- Will merge to 'next'.
722
+ Will merge to 'master'.
723
724
725
* jk/lore-is-the-archive (2019-11-30) 2 commits
880
- - doc: replace public-inbox links with lore.kernel.org
881
- - doc: recommend lore.kernel.org over public-inbox.org
726
+ (merged to 'next' on 2019-12-02 at 603a2b0105)
727
+ + doc: replace public-inbox links with lore.kernel.org
728
+ + doc: recommend lore.kernel.org over public-inbox.org
729
+ (this branch is used by dl/lore-is-the-archive.)
730
731
Doc update for the mailing list archiving and nntp service.
732
885
- Will merge to 'next'.
733
+ Will merge to 'master'.
734
735
736
* jk/send-pack-check-negative-with-quick (2019-11-30) 1 commit
889
- - send-pack: use OBJECT_INFO_QUICK to check negative objects
737
+ (merged to 'next' on 2019-12-02 at 2b01214601)
738
+ + send-pack: use OBJECT_INFO_QUICK to check negative objects
739
740
Performance tweak on "git push" into a repository with many refs
741
that point at objects we have never heard of.
742
894
- Will merge to 'next'.
743
+ Will merge to 'master'.
744
745
746
* sg/osx-force-gcc-9 (2019-11-29) 1 commit
898
- - ci: build Git with GCC 9 in the 'osx-gcc' build job
747
+ (merged to 'next' on 2019-12-02 at 4916f80e5e)
748
+ + ci: build Git with GCC 9 in the 'osx-gcc' build job
749
750
TravisCI update.
751
902
- Will merge to 'next'.
903
-
904
---------------------------------------------------
905
-[Stalled]
906
-
907
-* po/bundle-doc-clonable (2019-10-21) 1 commit
908
- - Doc: Bundle file usage
909
-
910
- Doc update.
911
-
912
- Expecting a reroll.
913
-
914
-
915
-* js/advise-rebase-skip (2019-10-23) 3 commits
916
- - commit: give correct advice for empty commit during a rebase
917
- - sequencer: export the function to get the path of `.git/rebase-merge/`
918
- - cherry-pick: add test for `--skip` advice in `git commit`
919
-
920
- The logic used in "git commit" to give hints and errors depending
921
- on what operation was in progress learned to distinguish rebase and
922
- cherry-pick better.
923
-
924
- Kicked back to 'pu'.
925
- cf. <nycvar.QRO.7.76.6.1911081508560.46@tvgsbejvaqbjf.bet>
926
-
927
-
928
-* vn/reset-deleted-ita (2019-07-26) 1 commit
929
- - reset: unstage empty deleted ita files
930
-
931
- "git reset HEAD [<pathspec>]" did not reset an empty file that was
932
- added with the intent-to-add bit.
933
-
934
- Expecting a reroll.
935
-
936
-
937
-* jn/unknown-index-extensions (2018-11-21) 2 commits
938
- - index: offer advice for unknown index extensions
939
- - index: do not warn about unrecognized extensions
940
-
941
- A bit too alarming warning given when unknown index extensions
942
- exist is getting revamped.
943
-
944
- Expecting a reroll.
945
-
946
-
947
-* jc/format-patch-delay-message-id (2019-04-05) 1 commit
948
- - format-patch: move message-id and related headers to the end
949
-
950
- The location "git format-patch --thread" adds the Message-Id:
951
- header in the series of header fields has been moved down, which
952
- may help working around a suspected bug in GMail MSA, reported at
953
- <CAHk-=whP1stFZNAaJiMi5eZ9rj0MRt20Y_yHVczZPH+O01d+sA@mail.gmail.com>
954
-
955
- Waiting for feedback to see if it truly helps.
956
- Needs tests.
957
-
958
-
959
-* js/protocol-advertise-multi (2018-12-28) 1 commit
960
- - protocol: advertise multiple supported versions
961
-
962
- The transport layer has been updated so that the protocol version
963
- used can be negotiated between the parties, by the initiator
964
- listing the protocol versions it is willing to talk, and the other
965
- side choosing from one of them.
966
-
967
- Expecting a reroll.
968
- cf. <CANq=j3u-zdb_FvNJGPCmygNMScseav63GhVvBX3NcVS4f7TejA@mail.gmail.com>
969
-
970
-
971
-* mk/use-size-t-in-zlib (2018-10-15) 1 commit
972
- - zlib.c: use size_t for size
973
-
974
- The wrapper to call into zlib followed our long tradition to use
975
- "unsigned long" for sizes of regions in memory, which have been
976
- updated to use "size_t".
977
-
978
---------------------------------------------------
979
-[Cooking]
980
-
981
-* jh/userdiff-python-async (2019-11-20) 1 commit
982
- (merged to 'next' on 2019-11-25 at 74e4a4ed9f)
983
- + userdiff: support Python async functions
984
-
985
- The userdiff machinery has been taught that "async def" is another
986
- way to begin a "function" in Python.
987
-
752
Will merge to 'master'.
753
754
755
* dl/pretty-reference (2019-11-20) 11 commits
992
- - SubmittingPatches: use `--pretty=reference`
993
- - pretty: implement 'reference' format
994
- - pretty: add struct cmt_fmt_map::default_date_mode_type
995
- - pretty: provide short date format
996
- - t4205: cover `git log --reflog -z` blindspot
997
- - pretty.c: inline initalize format_context
998
- - revision: make get_revision_mark() return const pointer
999
- - completion: complete `tformat:` pretty format
1000
- - SubmittingPatches: remove dq from commit reference
1001
- - pretty-formats.txt: use generic terms for hash
1002
- - SubmittingPatches: use generic terms for hash
756
+ (merged to 'next' on 2019-12-05 at 0972133ad2)
757
+ + SubmittingPatches: use `--pretty=reference`
758
+ + pretty: implement 'reference' format
759
+ + pretty: add struct cmt_fmt_map::default_date_mode_type
760
+ + pretty: provide short date format
761
+ + t4205: cover `git log --reflog -z` blindspot
762
+ + pretty.c: inline initalize format_context
763
+ + revision: make get_revision_mark() return const pointer
764
+ + completion: complete `tformat:` pretty format
765
+ + SubmittingPatches: remove dq from commit reference
766
+ + pretty-formats.txt: use generic terms for hash
767
+ + SubmittingPatches: use generic terms for hash
768
769
"git log" family learned "--pretty=ref" that gives the name of a
770
commit in the format that is often used to refer to it in log
771
messages.
772
1008
- Will merge to 'next'.
773
+ Will merge to 'master'.
774
775
776
* hw/doc-in-header (2019-11-18) 20 commits
@@ -1032,47 +797,20 @@ of the repositories listed at
797
798
799
* js/builtin-add-i-cmds (2019-12-01) 9 commits
1035
- - built-in add -i: offer the `quit` command
1036
- - built-in add -i: re-implement the `diff` command
1037
- - built-in add -i: implement the `patch` command
1038
- - built-in add -i: re-implement `add-untracked` in C
1039
- - built-in add -i: re-implement `revert` in C
1040
- - built-in add -i: implement the `update` command
1041
- - built-in add -i: prepare for multi-selection commands
1042
- - built-in add -i: allow filtering the modified files list
1043
- - add-interactive: make sure to release `rev.prune_data`
1044
- (this branch uses js/builtin-add-i.)
800
+ (merged to 'next' on 2019-12-05 at 30cfe5c15d)
801
+ + built-in add -i: offer the `quit` command
802
+ + built-in add -i: re-implement the `diff` command
803
+ + built-in add -i: implement the `patch` command
804
+ + built-in add -i: re-implement `add-untracked` in C
805
+ + built-in add -i: re-implement `revert` in C
806
+ + built-in add -i: implement the `update` command
807
+ + built-in add -i: prepare for multi-selection commands
808
+ + built-in add -i: allow filtering the modified files list
809
+ + add-interactive: make sure to release `rev.prune_data`
810
811
"git add -i" that is getting rewritten in C has been extended to
812
cover subcommands other than the "patch".
813
1049
- Will merge to 'next'.
1050
-
1051
-
1052
-* js/rebase-r-safer-label (2019-11-18) 2 commits
1053
- (merged to 'next' on 2019-11-21 at 791d51b0d7)
1054
- + rebase -r: let `label` generate safer labels
1055
- + rebase-merges: move labels' whitespace mangling into `label_oid()`
1056
- (this branch is used by dd/rebase-merge-reserves-onto-label.)
1057
-
1058
- A label used in the todo list that are generated by "git rebase
1059
- --rebase-merges" is used as a part of a refname; the logic to come
1060
- up with the label has been tightened to avoid names that cannot be
1061
- used as such.
1062
-
1063
- Will merge to 'master'.
1064
-
1065
-
1066
-* dd/rebase-merge-reserves-onto-label (2019-11-20) 1 commit
1067
- (merged to 'next' on 2019-11-21 at a6e4e5af0a)
1068
- + sequencer: handle rebase-merges for "onto" message
1069
- (this branch uses js/rebase-r-safer-label.)
1070
-
1071
- The logic to avoid duplicate label names generated by "git rebase
1072
- --rebase-merges" forgot that the machinery itself uses "onto" as a
1073
- label name, which must be avoided by auto-generated labels, which
1074
- has been corrected.
1075
-
814
Will merge to 'master'.
815
816
@@ -1128,6 +866,7 @@ of the repositories listed at
866
+ doc: reset: synchronize <pathspec> description
867
+ pathspec: add new function to parse file
868
+ parse-options.h: add new options `--pathspec-from-file`, `--pathspec-file-nul`
869
+ (this branch is used by am/pathspec-f-f-checkout and es/pathspec-f-f-grep.)
870
871
A few commands learned to take the pathspec from the
872
standard input or a named file, instead of taking it as the command
@@ -1136,65 +875,6 @@ of the repositories listed at
875
Will merge to 'master'.
876
877
1139
-* ec/fetch-mark-common-refs-trace2 (2019-11-20) 1 commit
1140
- (merged to 'next' on 2019-11-25 at c2b8e1546c)
1141
- + fetch: add trace2 instrumentation
1142
-
1143
- Trace2 annotation.
1144
-
1145
- Will merge to 'master'.
1146
-
1147
-
1148
-* ep/guard-kset-tar-headers (2019-11-07) 1 commit
1149
- (merged to 'next' on 2019-11-21 at 6c9574b01d)
1150
- + kset.h, tar.h: add missing header guard to prevent multiple inclusion
1151
-
1152
- Code clean-up.
1153
-
1154
- Will merge to 'master'.
1155
-
1156
-
1157
-* dl/t5520-cleanup (2019-11-21) 15 commits
1158
- (merged to 'next' on 2019-11-27 at 523d920acf)
1159
- + t5520: replace `! git` with `test_must_fail git`
1160
- + t5520: remove redundant lines in test cases
1161
- + t5520: replace $(cat ...) comparison with test_cmp
1162
- + t5520: don't put git in upstream of pipe
1163
- + t5520: test single-line files by git with test_cmp
1164
- + t5520: use test_cmp_rev where possible
1165
- + t5520: replace test -{n,z} with test-lib functions
1166
- + t5520: use test_line_count where possible
1167
- + t5520: remove spaces after redirect operator
1168
- + t5520: replace test -f with test-lib functions
1169
- + t5520: let sed open its own input
1170
- + t5520: use sq for test case names
1171
- + t5520: improve test style
1172
- + t: teach test_cmp_rev to accept ! for not-equals
1173
- + t0000: test multiple local assignment
1174
-
1175
- Test cleanup.
1176
-
1177
- Will merge to 'master'.
1178
-
1179
-
1180
-* js/builtin-add-i (2019-11-18) 9 commits
1181
- (merged to 'next' on 2019-11-21 at caefa55b89)
1182
- + built-in add -i: implement the `help` command
1183
- + built-in add -i: use color in the main loop
1184
- + built-in add -i: support `?` (prompt help)
1185
- + built-in add -i: show unique prefixes of the commands
1186
- + built-in add -i: implement the main loop
1187
- + built-in add -i: color the header in the `status` command
1188
- + built-in add -i: implement the `status` command
1189
- + diff: export diffstat interface
1190
- + Start to implement a built-in version of `git add --interactive`
1191
- (this branch is used by js/builtin-add-i-cmds.)
1192
-
1193
- The beginning of rewriting "git add -i" in C.
1194
-
1195
- Will merge to 'master'.
1196
-
1197
-
878
* ra/rebase-i-more-options (2019-11-25) 7 commits
879
(merged to 'next' on 2019-11-27 at e8a6d1e281)
880
+ rebase -i: finishing touches to --reset-author-date
@@ -1212,11 +892,12 @@ of the repositories listed at
892
893
894
* dl/submodule-set-url (2019-10-30) 1 commit
1215
- - submodule: teach set-url subcommand
895
+ (merged to 'next' on 2019-12-05 at a164eccad1)
896
+ + submodule: teach set-url subcommand
897
898
"git submodule" learned a subcommand "set-url".
899
1219
- Will merge to 'next'.
900
+ Will merge to 'master'.
901
902
903
* ag/sequencer-todo-updates (2019-11-25) 5 commits
@@ -1226,6 +907,7 @@ of the repositories listed at
907
+ sequencer: move the code writing total_nr on the disk to a new function
908
+ sequencer: update `done_nr' when skipping commands in a todo list
909
+ sequencer: update `total_nr' when adding an item to a todo list
910
+ (this branch is used by ag/edit-todo-drop-check.)
911
912
Reduce unnecessary reading of state variables back from the disk
913
during sequencer operation.
@@ -1233,27 +915,6 @@ of the repositories listed at
915
Will merge to 'master'.
916
917
1236
-* mt/threaded-grep-in-object-store (2019-10-02) 11 commits
1237
- - grep: move driver pre-load out of critical section
1238
- - grep: re-enable threads in non-worktree case
1239
- - grep: protect packed_git [re-]initialization
1240
- - grep: allow submodule functions to run in parallel
1241
- - submodule-config: add skip_if_read option to repo_read_gitmodules()
1242
- - grep: replace grep_read_mutex by internal obj read lock
1243
- - object-store: allow threaded access to object reading
1244
- - replace-object: make replace operations thread-safe
1245
- - grep: fix racy calls in grep_objects()
1246
- - grep: fix race conditions at grep_submodule()
1247
- - grep: fix race conditions on userdiff calls
1248
-
1249
- Traditionally, we avoided threaded grep while searching in objects
1250
- (as opposed to files in the working tree) as accesses to the object
1251
- layer is not thread-safe. This limitation is getting lifted.
1252
-
1253
- Expecting a reroll.
1254
- cf. <CAHd-oW7UPSSExyLtfLMCObWogKrBOctYabrFrOdf9-4Q2PZmMg@mail.gmail.com>
1255
-
1256
-
918
* jk/packfile-reuse-cleanup (2019-10-23) 9 commits
919
(merged to 'next' on 2019-11-19 at 9683853939)
920
+ pack-objects: improve partial packfile reuse